The application proposes to: 1) develop a computer simulation of idealized drug dose (input) data and patient response measures (output); the doses are of opiate analgesic and/or tricyclic adjunctive drugs; the output is to be assessed by questionnaire instruments, soon to be tested; 2) analyze the results of an ongoing trail, comparing amitriptyline and desipramine adjunctive analgesia in cancer pain patients treated with opiates; and 3) conduct a prospective, blinded, randomized study of veniafaxine versus placebo adjunctive treatment (to opiate analgesia) in cancer pain patients; the opioid treatment regiment will first have been stabilized for each patient. The veniafaxine (or placebo, benztropine) dose will be titrated upward, over time, unless limited by side effects. Patients will be followed for ten weeks, their condition being scored at either daily, weekly or bi-weekly intervals by the defined questionnaire battery. Over this time period, not only will adjunct does be varied upward or downward, depending on the complaint of side effect, whereas the opiod dose will be varied in response to pain complaint.